CN115861781A - Commodity identification method and device - Google Patents
Commodity identification method and device Download PDFInfo
- Publication number
- CN115861781A CN115861781A CN202211672926.6A CN202211672926A CN115861781A CN 115861781 A CN115861781 A CN 115861781A CN 202211672926 A CN202211672926 A CN 202211672926A CN 115861781 A CN115861781 A CN 115861781A
- Authority
- CN
- China
- Prior art keywords
- commodity
- identified
- shelf
- category
- determining
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The application discloses a commodity identification method and a device, and the method comprises the following steps: determining commodity categories corresponding to commodities to be identified according to the display positions of the commodities to be identified; and based on the commodity features of the commodities to be identified, performing comparison retrieval in a commodity feature library corresponding to the commodity categories, and selecting the feature category with the largest similarity with the commodity features from the commodity feature library as the category of the commodities to be identified. According to the method and the device, the corresponding commodity types are determined according to the display positions of the commodities to be identified, so that the commodity types in the commodity feature library are reduced, and the commodity retrieval identification is performed by using the reduced commodity feature library, so that the resource cost and the time cost are saved, and the more accurate commodity identification is realized.
Description
Technical Field
The application belongs to the technical field of artificial intelligence, and particularly relates to a commodity identification method and device.
Background
With the rapid development of artificial intelligence, intelligent retail is rapidly developed under the promotion of the artificial intelligence. In order to realize efficient and intelligent commodity management, technologies based on the internet, big data, artificial intelligence and the like are comprehensively used for intelligent retail, and the existing business superman and convenience stores are enabled or upgraded to be more digital and intelligent.
The intelligent digital goods shelf is an important link in the intelligent retail business, and the intelligent management requirements of the intelligent digital goods shelf are met by the functions of intelligent detection, intelligent identification, goods shortage, display supervision and the like of goods on the goods shelf. The commodity identification is an important link in intelligent goods shelf management, is also a precondition for the application of goods shelf ranking analysis, goods shelf shortage analysis and the like, and is a core technical point for realizing the intelligent management of goods shelves. Therefore, accurate product identification plays a crucial role in intelligent management of shelves. Due to the various kinds of commodities, frequent updating of commodity packages, changes of actual scene environments and other influences, high-precision commodity identification is difficult greatly, massive data support is needed, and a precise commodity identification technology is needed.
In order to realize high-precision commodity identification, the most effective method at present is to search commodities in a graph searching mode, namely a system is provided with a feature vector database, distance calculation or similarity calculation is carried out between features on a target feature and the feature vector database during searching, and the category to which the feature with the highest similarity belongs is taken as the category of the target feature. However, due to the fact that the variety of commodities is large, resource consumption and time consumption of commodity identification are large, accuracy and efficiency cannot be achieved, and subsequent intelligent shelf management is affected.
Content of application
The embodiment of the application aims to provide a commodity identification method and a commodity identification device so as to overcome the defects of high resource consumption and time consumption of commodity identification in the prior art.
In order to solve the technical problem, the present application is implemented as follows:
in a first aspect, a method for identifying a commodity is provided, which includes the following steps:
determining commodity classes corresponding to commodities to be identified according to the display positions of the commodities to be identified;
and based on the commodity features of the commodities to be identified, performing comparison retrieval in a commodity feature library corresponding to the commodity categories, and selecting the feature category with the largest similarity with the commodity features from the commodity feature library as the category of the commodities to be identified.
In a second aspect, there is provided an article identification device comprising:
the determining module is used for determining the commodity class corresponding to the commodity to be identified according to the display position of the commodity to be identified;
and the retrieval module is used for carrying out comparison retrieval in the commodity feature library corresponding to the commodity category based on the commodity features of the commodity to be identified, and selecting the feature category with the largest similarity with the commodity features from the commodity feature library as the category of the commodity to be identified.
According to the method and the device, the corresponding commodity types are determined according to the display positions of the commodities to be identified, so that the commodity types in the commodity feature library are reduced, and the commodity retrieval identification is performed by using the reduced commodity feature library, so that the resource cost and the time cost are saved, and the more accurate commodity identification is realized.
Drawings
Fig. 1 is a flowchart of a method for identifying a commodity according to an embodiment of the present application;
FIG. 2 is a schematic view of a shelf layer for holding goods according to an embodiment of the present disclosure;
fig. 3 is a schematic structural diagram of a product identification device according to an embodiment of the present application.
Detailed Description
The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some, but not all, embodiments of the present application. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
In the embodiment of the application, all commodity pictures needing to be identified in the business surpass can be collected, the commodity identification feature library is constructed, all commodity features in the whole commodity feature library are used for feature comparison during commodity identification, and finally the category to which the feature with the highest similarity belongs is taken as the category of the commodity to be identified. The commodity identification technology uses the feature library constructed by all commodities to carry out commodity feature comparison and retrieval, millions or even millions of commodity features exist in the whole feature library, and accurate commodity identification becomes extremely difficult due to massive commodity features. Meanwhile, comparing and searching with massive features will consume a lot of resources and time cost, and real-time performance cannot be achieved.
Therefore, the present embodiment provides a method for identifying a product based on shelf information, which uses shelf product information (or more detailed shelf product standard display information) provided by a merchant, and can reduce the feature class of the feature library to be compared and searched to the class range of the product placed on the shelf according to the shelf product information. If the quotient exceeds the standard display information of the commodity on the shelf, the characteristic types of the characteristic library to be compared and searched can be reduced to a smaller commodity type range, and therefore the number of the characteristics in the characteristic library is reduced. The commodity information or standard display information of the commodities on the goods shelf is used for acquiring the types and categories of the commodities on the goods shelf and even the placing sequence of the commodities on the goods shelf, and then the commodity feature library is reduced based on the commodity types or the category information, so that the commodity features irrelevant to the commodities to be identified in the commodity feature library are reduced as much as possible.
In actual situations, in order to manage the commodities on the shelves, commodities of the same category are usually placed on the same shelf (or the same layer of a certain shelf), that is, the category of the commodity on one shelf is fixed (or known). Further, some merchants place the products according to the standard display of the goods on shelves prescribed by the brand merchants, depending on the needs of the brand merchants, in which case the standard display may provide more detailed information on the placement of the products. According to the commodity identification method provided by the embodiment of the application, the commodity feature library for retrieval and comparison is firstly reduced within a certain range before commodity retrieval according to the commodity information or the standard display information of the commodity on the goods shelf, and then more accurate commodity retrieval and identification are carried out in the reduced commodity feature library, so that the commodity feature quantity for commodity retrieval and comparison can be reduced, the accuracy of commodity identification is improved, the resource consumption and the time consumption during commodity retrieval can be reduced, and the commodity identification is really fast and accurate.
A commodity identification method provided in the embodiments of the present application is described in detail below with reference to the accompanying drawings through specific embodiments and application scenarios thereof.
As shown in fig. 1, a flowchart of a method for identifying a commodity according to an embodiment of the present application is provided, where the method includes the following steps:
Specifically, the commodity category corresponding to the shelf where the commodity to be identified is located may be determined as the commodity category corresponding to the commodity to be identified; wherein all of the items on the shelf belong to the item class. The method also comprises the steps of determining a first commodity category corresponding to the display position of the commodity to be identified on the shelf and a second commodity category corresponding to the position adjacent to the display position according to standard display information of the shelf commodities; determining the commodity class corresponding to the commodity to be identified according to the commodity classes corresponding to the first commodity class and the second commodity class; wherein the shelf commodity standard display information is used to determine the category of commodity placed at each location in the shelf.
In this embodiment, the commodity categories corresponding to the first commodity category and the second commodity category may be determined as the commodity categories corresponding to the commodities to be identified.
And 102, based on the commodity features of the commodities to be identified, performing comparison retrieval in a commodity feature library corresponding to the commodity categories, and selecting the feature category with the largest similarity with the commodity features from the commodity feature library as the category of the commodities to be identified.
In this embodiment, before determining the commodity category corresponding to the commodity to be identified according to the display position of the commodity to be identified, the position of the shelf to which the commodity to be identified belongs on the shelf can be positioned according to the coordinate of the central point of the detection frame of the commodity to be identified; correspondingly, a first commodity category corresponding to the position of the grid to which the commodity to be identified belongs on the shelf and a second commodity category corresponding to the positions of a plurality of grids adjacent to the left and the right of the grid can be determined according to the standard display information of the goods on the shelf.
According to the method and the device, the corresponding commodity types are determined according to the display positions of the commodities to be identified, so that the commodity types in the commodity feature library are reduced, and the commodity retrieval identification is performed by using the reduced commodity feature library, so that the resource cost and the time cost are saved, and the more accurate commodity identification is realized.
In the embodiment of the application, the category and the number of the features in the commodity feature library to be compared and retrieved are reduced through shelf commodity information or shelf commodity standard display information provided by a merchant, so that commodities to be identified are compared with fewer features of the feature library during commodity identification, a more accurate commodity identification result is obtained, and less resource cost and time cost are consumed. Wherein the category of the goods represents a certain large category of goods, such as: beverages, cookies, washing and caring products, etc., and the category of the product represents the specific category of the product. The shelf grid represents a fixed area location on the shelf, as shown by the dashed box based shelf level in fig. 2.
In a specific implementation of the embodiment of the present application, when the shelf commodity information is provided by the merchant surpassing, the commodity identification method for narrowing the commodity feature library based on the shelf commodity information specifically implements the following steps:
(1) Assuming that there are 10 categories of commodities in the overall commodity feature library, such as A, B, C, D, E, F, G, H, I, J;
(2) For a commodity to be identified, the commodity class on the shelf where the commodity is located belongs to A or B;
(3) Narrowing down the product feature library to contain features of only two types of products A and B using the shelf product information (types A and B) provided in (2);
(4) Comparing and searching the characteristics of the commodities to be identified with the characteristics in the characteristic library with the reduced commodity category in the step (3);
(5) And selecting the characteristic category with the maximum similarity as the category of the commodity to be identified.
In another specific implementation of the embodiment of the present application, the method for identifying a commodity based on a commodity feature library with reduced standard display information of a commodity on a shelf specifically implements the following steps:
(1) The commodity category of each position of each layer of the shelf can be obtained according to the standard display information of the goods on the shelf. Assuming that the goods marked in the first layer of the shelf in fig. 2 are goods to be identified, the coordinates of the center point of the detection frame can be calculated according to the detection frame of the goods to be identified;
(2) The position of the shelf where the commodity detection frame belongs can be located according to the coordinates of the center point of the commodity detection frame calculated in the step (1), such as the shelf where the category A4 belongs in the figure 2;
(3) Selecting n grids on the left and right of the grid to which the commodity belongs to obtain commodity categories within the range of 2n +1 grids; assuming that n =1, selecting three grids A3, A4 and A5, and obtaining three commodity categories A3, A4 and A5;
(4) According to the 2n +1 commodity categories selected in (3), the categories of the features in the commodity feature library are reduced to 2n +1 categories;
(5) Comparing and searching the characteristics of the commodities to be identified with the characteristics in the characteristic library with the reduced commodity category in the step (4);
(6) And selecting the characteristic category with the maximum similarity as the category of the commodity to be identified.
Before commodity identification is carried out on commodities on a goods shelf, feature selection is carried out on a full-grade commodity feature library according to commodity information of the goods shelf or standard display information of the goods shelf, features related to the commodity class of the commodities on the goods shelf or the commodity class of the selected shed lattice are selected, the selected commodity features are used for subsequent commodity identification, the classes and the categories of the commodities in the commodity feature library can be greatly reduced, the number of features needing to be compared is reduced, and then the reduced feature library is used for commodity retrieval and identification, so that a large amount of resources and time cost can be saved, and more accurate commodity identification can be realized.
As shown in fig. 3, a schematic structural diagram of a product identification device provided in an embodiment of the present application includes:
the determining module 310 is configured to determine, according to the display position of the to-be-identified product, a product category corresponding to the to-be-identified product.
Specifically, the determining module 310 may be specifically configured to determine a commodity category corresponding to a shelf where a commodity to be identified is located as the commodity category corresponding to the commodity to be identified; wherein all the commodities on the shelf belong to the commodity class. The method can also be specifically used for determining a first commodity category corresponding to the display position of the commodity to be identified on the shelf and a second commodity category corresponding to the position adjacent to the display position according to the standard display information of the shelf commodities; determining the commodity class corresponding to the commodity to be identified according to the commodity classes corresponding to the first commodity class and the second commodity class; wherein the shelf commodity standard display information is used to determine the category of commodity placed at each location in the shelf.
In this embodiment, the determining module 310 is specifically configured to determine the commodity categories corresponding to the first commodity category and the second commodity category as the commodity categories corresponding to the commodities to be identified.
The retrieval module 320 is configured to perform comparison retrieval in the commodity feature library corresponding to the commodity category based on the commodity features of the commodity to be identified, and select a feature category with the largest similarity to the commodity features from the commodity feature library as the category of the commodity to be identified.
In this embodiment, the above apparatus further includes:
the positioning module is used for positioning the position of the goods to be identified on the shelf in the shed lattice according to the coordinates of the central point of the detection frame of the goods to be identified;
accordingly, the determining module 310 is specifically configured to determine, according to the standard display information of the goods on the shelf, a first goods category corresponding to the position of the shelf to which the goods to be identified belong on the shelf, and a second goods category corresponding to the positions of multiple shelves adjacent to the shelf on the left and right.
According to the method and the device, the corresponding commodity types are determined according to the display positions of the commodities to be identified, so that the commodity types in the commodity feature library are reduced, and the commodity retrieval identification is performed by using the reduced commodity feature library, so that the resource cost and the time cost are saved, and the more accurate commodity identification is realized.
It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrases "comprising a component of' 8230; \8230;" does not exclude the presence of another like element in a process, method, article, or apparatus that comprises the element.
Through the above description of the embodiments, those skilled in the art will clearly understand that the method of the above embodiments can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware, but in many cases, the former is a better implementation manner. Based on such understanding, the technical solutions of the present application may be embodied in the form of a software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk) and includes instructions for enabling a terminal (such as a mobile phone, a computer, a server, an air conditioner, or a network device) to execute the method according to the embodiments of the present application.
While the present embodiments have been described with reference to the accompanying drawings, it is to be understood that the invention is not limited to the precise embodiments described above, which are meant to be illustrative and not restrictive, and that various changes may be made therein by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims.
Claims (10)
1. A method of identifying a commodity, comprising the steps of:
determining commodity classes corresponding to commodities to be identified according to the display positions of the commodities to be identified;
and based on the commodity features of the commodities to be identified, performing comparison retrieval in a commodity feature library corresponding to the commodity categories, and selecting the feature category with the largest similarity with the commodity features from the commodity feature library as the category of the commodities to be identified.
2. The method according to claim 1, wherein the determining of the commodity category corresponding to the commodity to be identified according to the display position of the commodity to be identified specifically comprises:
determining the commodity class corresponding to the goods shelf where the commodity to be identified is located as the commodity class corresponding to the commodity to be identified;
wherein all of the items on the shelf belong to the item class.
3. The method according to claim 1, wherein the determining of the commodity class corresponding to the commodity to be identified according to the display position of the commodity to be identified specifically comprises:
determining a first commodity category corresponding to the display position of the commodity to be identified on a shelf and a second commodity category corresponding to the position adjacent to the display position according to standard display information of the shelf commodity;
determining the commodity class corresponding to the commodity to be identified according to the commodity classes corresponding to the first commodity class and the second commodity class;
wherein the shelf commodity standard display information is used to determine the category of commodity placed at each location in the shelf.
4. The method according to claim 3, wherein the determining, according to the first item category and the second item category, an item category corresponding to the item to be identified specifically comprises:
and determining the commodity classes corresponding to the first commodity class and the second commodity class as the commodity classes corresponding to the commodities to be identified.
5. The method of claim 3, wherein before determining the item class corresponding to the item to be identified according to the display position of the item to be identified, the method further comprises:
positioning the position of the goods to be identified on the shelf in the grid according to the coordinates of the central point of the detection frame of the goods to be identified;
the method for identifying the commodity type comprises the following steps of determining a first commodity type corresponding to a display position of a commodity to be identified on a shelf and a second commodity type corresponding to a position adjacent to the display position according to standard display information of the shelf commodity, and specifically comprises the following steps:
according to the standard display information of the goods on the shelf, a first goods category corresponding to the position 5 of the grids on the shelf to which the goods to be identified belong and a second goods category corresponding to the positions of a plurality of grids adjacent to the left and the right of the grids are determined.
6. An article identification device, comprising:
the determining module is used for determining the commodity class of 0 corresponding to the commodity to be identified according to the display position of the commodity to be identified;
and the retrieval module is used for carrying out comparison retrieval in the commodity feature library corresponding to the commodity category based on the commodity features of the commodity to be identified, and selecting the feature category with the largest similarity with the commodity features from the commodity feature library as the category of the commodity to be identified.
7. The apparatus of claim 6,
the determining module is specifically configured to determine a commodity category corresponding to a shelf on which a commodity to be identified is located as the commodity category corresponding to the commodity to be identified; wherein all the commodities on the shelf belong to the commodity class.
8. The apparatus of claim 6,
the determining module is specifically used for determining a first commodity category corresponding to the display position of the commodity to be identified on the 0 shelf and a second commodity category corresponding to the position adjacent to the display position according to the standard display information of the goods on the shelf; determining the commodity class corresponding to the commodity to be identified according to the commodity classes corresponding to the first commodity class and the second commodity class; wherein the shelf commodity standard display information is used to determine the category of commodity placed at each location in the shelf.
9. The apparatus of claim 8,
and 5, the determining module is specifically configured to determine the commodity categories corresponding to the first commodity category and the second commodity category as the commodity categories corresponding to the commodities to be identified.
10. The apparatus of claim 8, further comprising:
the positioning module is used for positioning the position of the goods to be identified on the shelf in the shed lattice according to the coordinates of the central point of the detection frame of the goods to be identified;
the determining module is specifically used for determining a first commodity category corresponding to the positions of the grids on the shelf to which the commodities to be identified belong and a second commodity category corresponding to the positions of a plurality of grids adjacent to the left and right of the grids according to the standard display information of the goods on the shelf.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211672926.6A CN115861781A (en) | 2022-12-26 | 2022-12-26 | Commodity identification method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211672926.6A CN115861781A (en) | 2022-12-26 | 2022-12-26 | Commodity identification method and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115861781A true CN115861781A (en) | 2023-03-28 |
Family
ID=85654697
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211672926.6A Pending CN115861781A (en) | 2022-12-26 | 2022-12-26 | Commodity identification method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115861781A (en) |
-
2022
- 2022-12-26 CN CN202211672926.6A patent/CN115861781A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106126630B (en) | A kind of collection of business object, searching method and device | |
| CN102214187B (en) | Complex event processing method and device | |
| CN102253936B (en) | Method for recording access of user to merchandise information, search method and server | |
| CN108320171A (en) | Hot item prediction technique, system and device | |
| CN109767150B (en) | Information pushing method and device | |
| CN111666275B (en) | Data processing method and device, electronic equipment and storage medium | |
| CN110909182A (en) | Multimedia resource searching method and device, computer equipment and storage medium | |
| CN107679103B (en) | Attribute analysis method and system for entity | |
| CN109214730A (en) | Information-pushing method and device | |
| CN111090822A (en) | Business object pushing method and device | |
| CN107911448A (en) | Content pushing method and device | |
| US20040002983A1 (en) | Method and system for detecting tables to be modified | |
| CN108665156B (en) | Supply chain selection evaluation method based on Markov chain under block chain | |
| CN110795613A (en) | Commodity searching method, device and system and electronic equipment | |
| CN116579804A (en) | Holiday commodity sales prediction method, holiday commodity sales prediction device and computer storage medium | |
| CN113627995A (en) | Commodity recommendation list updating method and device, equipment, medium and product thereof | |
| US8195703B1 (en) | System and method for storage of disparate items by a database | |
| CN115861781A (en) | Commodity identification method and device | |
| CN112749325A (en) | Training method and device for search ranking model, electronic equipment and computer medium | |
| CN115936875A (en) | Financial product form hanging processing method and device | |
| CN115423555A (en) | Commodity recommendation method and device, electronic equipment and storage medium | |
| CN112200644B (en) | Method and device for identifying fraudulent user, computer equipment and storage medium | |
| CN111598510B (en) | Order batching method suitable for intelligent warehousing system | |
| US10430775B1 (en) | Validation and lookup techniques for rule-based data categorization | |
| CN111027326A (en) | Commodity classification method, storage medium and electronic device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |